Influence of processing of full fat soya beans included in diets for piglets II. Digestibility and intestinal morphology

The objective of the current experiment was to investigate the effect of inclusion of differently micronized (Trial 1) and extruded (Trial 2) full fat soya beans (FFSB) on ileal and total tract apparent digestibility of FFSB in piglets. Histological examinations of the small intestine mucosa were also conducted. Data for coefficient of total tract apparent digestibility (CTTAD) revealed no differences for the effect of steam (S) or heat (H) in Trial 1 (P > 0.05). However S x H was significant for CCTAD for crude protein (CP) (P = 0.002). Full steam high cook (FSHC), full steam low cook (FSLC) and low steam high cook (LSHC) diets maintained similar villus height (VH) and no effect of S or H or S x H was observed at both jejunum and ileum, while only the S x T interaction was significant (P = 0.008) for crypt depth (CD). Trypsin inhibitor activity (TIA) content of the FFSB and the TIA daily intake influenced the VH/CD ratio (P < 0.05); values were reduced progressively from 3.52 for the FSHC diet to 2.86 for the LSLC. In Trial 2 CTTAD for dry matter (DM), gross energy (GE) and CP were not related to the different extrusion temperatures investigated (P > 0.05). Coefficients of ileal apparent digestibility (CIAD) for the 150degreesC extruded FFSB (E-FFSB) diets were consistently lower for ileal DM, GE and EE, and they were significantly related to the different extrusion temperatures investigated (P = 0.042, P = 0.050 and P = 0.028, respectively) compared to 90degreesC E-FFSB diets. VH was consistently higher for the low TIA diet (150degreesC E-FFSB) and CD was correlated to the FFSB TIA levels and TIA daily intake at the same site (P < 0.05). When digestibility coefficients were regressed against the levels of TIA in FFSB no relationships were observed (P > 0.05) for both trials. Thermal processing of the FFSB did not therefore affect digestibility while small intestinal structure, as expressed by VH and CD values, was not affected to any significant extent. Villus atrophy was not related to feed intake. However villi dimensions were associated with the dietary daily TIA intake. Combined results from both trials suggest that adequate inactivation of the anti-nutritional factors (ANFs) resulted in an increased VH and a lower TIA intake was associated with a higher mean VH. It is thus intake and not dietary concentration of TIA which is the major factor influencing responses. (C) 2004 Elsevier B.V. All rights reserved.